Research On Energy Management Strategy Of Hybrid Electric Vehicle Longitudinal Driving Based On Traffic Scene

Plug-in hybrid electric vehicles(PHEVs)are an important direction for the development of today’s automobiles because they can combine the advantages of longer electric driving range and better energy economy.As the complexity of road conditions increases,the energy economy of PHEVs is greatly influenced by traffic factors,and the intelligent and optimal control of PHEVs under Telematics technology plays a key role in improving their energy economy.In this paper,we take the plug-in hybrid electric vehicle of P2 configuration as the research object,aim at improving the energy economy of the whole vehicle,and use MATLAB/Simulink as the simulation platform to study the longitudinal driving energy management strategy of hybrid electric vehicle based on traffic scenarios,and the main research contents are as follows:(1)The system structure of the hybrid vehicle is introduced,the specific operation mode is analyzed,and the physical model of the vehicle is established in MATLAB/Simulink based on the core power system components and the vehicle dynamics model,and the simulation verification platform of the energy management control strategy is established.(2)Apply VISSIM simulation software to establish longitudinal traffic scenarios at intersections under urban conditions,simulate microscopic traffic road networks,and carry out vehicle-vehicle communication and vehicle-roadside device communication to obtain vehicle network information and obtain real-time traffic data to provide data sources for vehicle longitudinal travel characteristics prediction.Using deep learning to train the three typical working conditions data in the traffic model,get the vehicle longitudinal driving characteristics prediction model respectively,and obtain the predicted vehicle speed under different prediction steps,so as to lay the foundation for the subsequent research on the longitudinal driving energy management strategy of hybrid vehicles based on traffic scenes.(3)With the optimal energy consumption economy as the goal,the target threshold of the battery state of charge(SOC)at the time of mode switching is calculated by genetic algorithm,and the specific power source how to torque distribution strategy when the vehicle is running is formulated based on the transient optimization algorithm,and the energy consumption economy of this energy management strategy is compared and studied with the logical threshold energy management strategy,which proves that the energy consumption economy of this strategy is greatly improved.Secondly,considering that the real-time optimal energy management strategy is easy to fall into the partial optimum and cannot obtain the global optimum of energy consumption economy,the minimum value principle is applied to establish the energy cost function of the whole vehicle,and the dynamic planning algorithm is used to solve the torque distribution under the predicted speed curve,and the rolling optimization is used to continuously update the control process to achieve the global optimal energy management.Simulation experiments are conducted,and by comparing the DP strategy with the real-time optimization strategy,it is proved that the DP-based strategy can better improve the energy economy of the whole vehicle.(4)In order to verify the studied hybrid vehicle longitudinal driving characteristics prediction model and PHEV energy management strategy,a virtual traffic scenario is constructed with a real experimental roadway as the background,traffic information is collected to establish the prediction model,and the feasibility of the vehicle longitudinal driving characteristics prediction model based on deep learning is verified.Then the optimal energy economy is calculated according to the energy management strategy established in the previous paper,and the simulation results of the researched strategy are comparatively analyzed with the results of the energy management strategy based on logical threshold values,as well as the comparison of the 100 km fuel consumption under NEDC conditions,which verifies the effectiveness of the longitudinal driving energy management strategy of hybrid electric vehicles based on traffic scenarios proposed in this paper.